The Coloured Noise Expansion and Parameter Estimation of Diffusion Processes

Size: px
Start display at page:

Download "The Coloured Noise Expansion and Parameter Estimation of Diffusion Processes"

Transcription

1 The Coloured Noise Expansion and Parameter Estimation of Diffusion Processes Simon M.J. Lyons School of Informatics University of Edinburgh Crichton Street, Edinburgh, EH8 9AB Simo Särkkä Aalto University Department of Biomedical Engineering and Computational Science Rakentajanaukio, 5 Espoo simo.sarkka@aalto.fi Amos J. Storkey School of Informatics University of Edinburgh Crichton Street, Edinburgh, EH8 9AB a.storkey@ed.ac.uk Abstract Stochastic differential equations SDE are a natural tool for modelling systems that are inherently noisy or contain uncertainties that can be modelled as stochastic processes. Crucial to the process of using SDE to build mathematical models is the ability to estimate parameters of those models from observed data. Over the past few decades, significant progress has been made on this problem, but we are still far from having a definitive solution. We describe a novel method of approximating a diffusion process that we show to be useful in Markov chain Monte-Carlo MCMC inference algorithms. We take the white noise that drives a diffusion process and decompose it into two terms. The first is a coloured noise term that can be deterministically controlled by a set of auxilliary variables. The second term is small and enables us to form a linear Gaussian small noise approximation. The decomposition allows us to take a diffusion process of interest and cast it in a form that is amenable to sampling by MCMC methods. We explain why many state-of-the-art inference methods fail on highly nonlinear inference problems, and we demonstrate experimentally that our method performs well in such situations. Our results show that this method is a promising new tool for use in inference and parameter estimation problems. Introduction Diffusion processes are a flexible and useful tool in stochastic modelling. Many important real world systems are currently modelled and best understood in terms of stochastic differential equations in general and diffusions in particular. Diffusions have been used to model prices of financial instruments [], chemical reactions [], firing patterns of individual neurons [3], weather patterns [4] and fmri data [5, 6, 7] among many other phenomena. The analysis of diffusions dates back to Feller and Kolmogorov, who studied them as the scaling limits of certain Markov processes see [8]. The theory of diffusion processes was revolutionised by Itô, who interpreted a diffusion process as the solution to a stochastic differential equation [9, ]. This viewpoint allows one to see a diffusion process as the randomised counterpart of an ordinary differential equation. One can argue that stochastic differential equations are the natural

2 tool for modelling continuously evolving systems of real valued quantities that are subject to noise or stochastic influences. The classical approach to mathematical modelling starts with a set of equations that describe the evolution of a system of interest. These equations are goverened by a set of input parameters for example particle masses, reaction rates, or more general constants of proportionality that determine the behaviour of the system. For practical purposes, it is of considerable interest to solve the inverse problem. Given the output of some system, what can be said about the parameters that govern it? In the present setting, we observe data which we hypothesize are generated by a diffusion. We would like to know what the nature of this diffusion is. For example, we may begin with a parametric model of a physical system, with a prior distribution over the parameters. In principle, one can apply Bayes theorem to deduce the posterior distribution. In practice, this is computationally prohibitive: it is necessary to solve a partial differential equation known as the Fokker-Planck equation see [] in order to find the transition density of the diffusion of interest. This solution is rarely available in closed form, and must be computed numerically. In this paper, we propose a novel approximation for a nonlinear diffusion process X. One heuristic way of thinking about a diffusion is as an ordinary differential equation that is perturbed by white noise. We demonstrate that one can replace the white noise by a coloured approximation without inducing much error. The nature of the coloured noise expansion method enables us to control the behaviour of the diffusion over various length-scales. This allows us to produce samples from the diffusion process that are consistent with observed data. We use these samples in a Markov chain Monte-Carlo MCMC inference algorithm. The main contributions of this paper are: Novel development of a method for sampling from the time-t marginal distribution of a diffusion process based on a coloured approximation of white noise. Demonstration that this approximation is a powerful and scalable tool for making parameter estimation feasible for general diffusions at minimal cost. The paper is structured as follows: in Section, we describe the structure of our problem. In Section 3 we conduct a brief survey of existing approaches to the problem. In Section 4, we discuss the coloured noise expansion and its use in controlling the behaviour of a diffusion process. Our inference algorithm is described in Section 5. We describe some numerical experiments in Section 6, and future work is discussed in Section 7. Parametric Diffusion Processes In this section we develop the basic notation and formalism for the diffusion processes used in this work. First, we assume our data are generated by observing a k-dimensional diffusion processes with dynamics dx t = a θ X t dt + B θ dw t, X px, where the initial condition is drawn from some known distribution. Observations are assumed to occur at times t,..., t n, with t i t i := T i. We require that a θ : IR k IR k is sufficiently regular to guarantee the existence of a unique strong solution to, and we assume B θ IR k d. Both terms depend on a set of potentially unknown parameters θ IR d θ. We impose a prior distribution pθ on the parameters. The driving noise W is a d-dimensional Brownian motion, and the equation is interpreted in the Itô sense. Observations are subject to independent Gaussian perturbations centered at the true value of X. That is, Y ti = X ti + ɛ ti, ɛ ti N, Σ i We use the notation X to refer to the entire sample path of the diffusion, and X t to denote the value of the process at time t. We will also employ the shorthand Y :n = {Y t,..., Y tn }. Many systems can be modelled using the form. Such systems are particularly relevant in physics and natural sciences. In situations where this is not explicitly the case, one can often hope to reduce a diffusion to this form via the Lamperti transform. One can almost always accomplish this in the univariate case, but the multivariate setting is somewhat more involved. Aït-Sahalia [] characterises the set of multivariate diffusions to which this transform can be applied.

3 3 Background Work Most approaches to parameter estimation of diffusion processes rely on the Monte-Carlo approximation. Beskos et al. [3] [4] employ a method based on rejection sampling to estimate parameters without introducing any discretisation error. Golightly and Wilkinson [5] extend the work of Chib et al. [6] and Durham and Gallant [7] to construct a Gibbs sampler that can be applied to the parameter estimation problem. Roughly speaking, Gibbs samplers that exist in the literature alternate between drawing samples from some representation of the diffusion process X conditional on parameters θ, and samples from θ conditional on the current sample path of X. Note that draws from X must be consistent with the observations Y :n. The usual approach to the consistency issue is to make a proposal by conditioning a linear diffusion to hit some neighbourhood of the observation Y k, then to make a correction via a rejection sampling [8] or a Metropolis-Hastings [6] step. However, as the inter-observation time grows, the qualitative difference between linear and nonlinear diffusions gets progressively more pronounced, and the rate of rejection grows accordingly. Figure shows the disparity between a sample from a nonlinear process and a sample from the linear proposal. One can see that the target sample path is constrained to stay near the mode γ =.5, whereas the proposal can move more freely. One should expect to make many proposals before finding one that behaves like a typical draw from the true process. Xt 3 Nonlinear sample path and proposal 3 t a Xt Sample path with noisy observations t Figure : a Sample path of a double well process see equation 8 with α =, γ =.5, B = blue line. Current Gibbs samplers use linear proposals dashed red line with a rejection step to draw conditioned nonlinear paths. In this case, the behaviour of the proposal is very different to that of the target, and the rate of rejection is high. b Sample path of a double well process solid blue line with noisy observations red dots. We use this as an initial dataset on which to test our algorithm. Parameters are α =, γ =, B =. Observation errors have variance Σ =.5. For low-dimensional inference problems, algorithms that employ sequential Monte-Carlo SMC methods [9] [] typically yield good results. However, unlike the Gibbs samplers mentioned above, SMC-based methods often do not scale well with dimension. The number of particles that one needs to maintain a given accuracy is known to scale exponentially with the dimension of the problem []. Aït-Sahalia [, ] uses a deterministic technique based on Edgeworth expansions to approximate the transition density. Other approaches include variational methods [3, 4] that can compute continuous time Gaussian process approximations to more general stochastic differential systems, as well as various non-linear Kalman filtering and smoothing based approximations [5, 6, 7]. 4 Coloured Noise Expansions and Brownian Motion We now introduce a method of approximating a nonlinear diffusion that allows us to gain a considerable amount of control over the behaviour of the process. Similar methods have been used b 3

4 for stratified sampling of diffusion processes [8] and the solution of stochastic partial differential equations [9]. One of the major challenges of using MCMC methods for parameter estimation in the present context is that it is typically very difficult to draw samples from a diffusion process conditional on observed data. If one only knows the initial condition of a diffusion, then it is straightforward to simulate a sample path of the process. However, simulating a sample path conditional on both initial and final conditions is a challenging problem. Our approximation separates the diffusion process X into the sum of a linear and nonlinear component. The linear component of the sum allows us to condition the approximation to fit observed data more easily than in conventional methods. On the other hand, the nonlinear component captures the gross variation of a typical sample path. In this section, we fix a generic time interval [, T ], though one can apply the same derivation for any given interval T i = t i t i. Heuristically, one can think of the random process that drives the process defined in equation as white noise. In our approximation, we project this white noise into an N-dimensional subspace of L [, T ], the Hilbert space of square-integrable functions defined on the interval [, T ]. This gives a coloured noise process that approaches white noise asymptotically as N. The coloured noise process is then used to drive an approximation of. We can choose the space into which to project the white noise in such a way that we will gain some control over its behaviour. This is analagous to the way that Fourier analysis allows us to manipulate properties of signals Recall that a standard Brownian motion on the interval [, T ] is a one-dimentional Gaussian process with zero mean and covariance function ks, t = min{s, t}. By definition of the Itô integral, we can write W t = t dw s = T I [,t] sdw s. 3 Suppose {φ i } i is an orthonormal basis of L [, T ]. We can interpret the indicator function in 3 as an element of L [, T ] and expand it in terms of the basis functions as follows: t I [,t] s = I [,t], φ i φ i s = φ i udu φ i s. 4 i= Substituting 4 into 3, we see that T t W t = φ i sdw s φ i udu. 5 i= We will employ the shorthand Z i = T φ isdw s. Since the functions {φ i } are deterministic and orthonormal, we know from standard results of Itô calculus that the random variables {Z i } are i.i.d standard normal. The infinite series in equation 5 can be truncated after N terms to derive an approximation, Ŵ t of Brownian motion. Taking the derivative with respect to time, the result is a coloured approximation of white noise, taking the form dŵt dt = i= N Z i φ i t. 6 i= The multivariate approximation is similar. We seperate a d-dimensional Brownian motion into onedimensional components and decompose the individual components as in 6. In principle, one can choose a different value of N for each component of the Brownian motion, but for ease of exposition we do not do so here. We can substitute this approximation into equation, which gives dx NL t dt = a θ X NL t + B θ N Φ i Z i, i= X NL px, 7 where Φ i is the diagonal d d matrix with entries φ i,..., φ id, and Z i = Z i,..., Z id. This derivation is useful because equation 7 gives us an alternative to the Euler-Maruyama discretisation for sampling approximately from the time-t marginal distribution of a diffusion process. We 4

5 draw coefficients Z ij from a standard normal distribution, and solve the appropriate vector-valued ordinary differential equation. While the Euler discretisation is the de facto standard method for numerical approximation of SDE, other methods do exist. Kloeden and Platen [3] discuss higher order methods such as the stochastic Runge-Kutta scheme [3]. In the Euler-Maruyama approximation, one discretises the driving Brownian motion into increments W ti W ti = T i Z i. One must typically employ a fine discretisation to get a good approximation to the true diffusion process. Empirically, we find that one needs far fewer Gaussian inputs Z i for an accurate representation of X T using the coloured noise approximation. This more parsimonious representation has advantages. For example, Corlay and Pages [8] employ related ideas to conduct stratified sampling of a diffusion process. The coefficients Z i are also more amenable to interpretation than the Gaussian increments in the Euler-Maruyama expansion. Suppose we have a one-dimensional process in which we use the Fourier cosine basis φ k t = /T cosk πt/t. 8 If we change Z while holding the other coefficients fixed, we will typically see a change in the large-scale behaviour of the path. On the other hand, a change in Z N will typically result in a change to the small-scale oscillations in the path. The seperation of behaviours across coefficients gives us a means to obtain fine-grained control over the behaviour of a diffusion process within a Metropolis-Hastings algorithm. We can improve our approximation by attempting to correct for the fact that we truncated the sum in equation 6. Instead of simply discarding the terms Z i Φ i for i > N, we attempt to account for their effect as follows. We assume the existence of some correction process X C such that X = X NL + X C. We know that the dynamics of X satisfy dx t = a θ X NL t + X C t dt + Bθ dw t. 9 Taylor expanding the drift term around X NL, we see that to first order, dx t a θ X NL t + Ja X NL t X C t dt + B θ dw t = a θ X NL t + Bθ dŵt dt + J a X NL t X C t dt + B θ dw t dŵt. Here, J a x is the Jacobian matrix of the function a evaluated at x. This motivates the use of a linear time-dependent approximation to the correction process. We will refer to this linear approximation as X L. The dynamics of X L satisfy dx L t = J a X NL t X L t dt + B θ dr t, X L =, where the driving noise is the residual term R = W Ŵ. Conditional on XNL, X L is a linear Gaussian process, and equation can be solved in semi-closed form. First, we compute a numerical approximation to the solution of the homogenous matrix-valued equation d dt Ψt = J ax NL t Ψt, Ψ = I n. One can compute Ψ t in a similar fashion via the relationship dψ /dt = Ψ dψ/dtψ. We then have X L t = Ψt = Ψt t t Ψu BdR u Ψu BdW u N i= t Ψt Ψu BΦ i udu Z i. 3 5

6 It follows that X L has mean and covariance s t ks, t = Ψs Ψu BB Ψ u du Ψ t N s Ψs i= t Ψu BΦ i udu Ψu BΦ i udu Ψ t. 4 The process X NL is designed to capture the most significant nonlinear features of the original diffusion X, while the linear process X L corrects for the truncation of the sum 6, and can be understood using tools from the theory of Gaussian processes. One can think of the linear term as the result of a small-noise expansion about the nonlinear trajectory. Small-noise techniques have been applied to diffusions in the past [], but the method described above has the advantage of being inherently nonlinear. In the supplement to this paper, we show that ˆX = X NL + X L converges to X in L [, T ] as N under the assumption that a is Lipschitz continuous. If the drift function is linear, then ˆX = X regardless of the choice of N. 5 Parameter Estimation In this section, we describe a novel modification of the Gibbs sampler that does not suffer the drawbacks of the linear proposal strategy. In Section 6, we demonstrate that for highly nonlinear problems it will perform significantly better than standard methods because of the nonlinear component of our approximation. Suppose for now that we make a single noiseless observation at time t = T for ease of notation, we will assume that observations are uniformly spaced through time with t i+ t i = T, though this is not necessary. Our aim is to sample from the posterior distribution p θ, Z :N X NL + X L = Y N Y X NL, k T, T N Z :N pθ. 5 We adopt the convention that N µ, Σ represents the normal distribution with mean µ and covariance Σ, whereas N represents the standard normal distribution. Note that we have left dependence of k on Z and θ implicit. The right-hand side of this expression allows us to evaluate the posterior up to proportionality; hence it can be targeted with a Metropolis-Hastings sampler. With multiple observations, the situation is similar. However, we now have a set of Gaussian inputs Z i for each transition ˆX i ˆX i. If we attempt to update θ and {Z i } i n all at once, the rate of rejection will be unacceptably high. For this reason, we update each Z i in turn, holding θ and the other Gaussian inputs fixed. We draw Z i from the proposal distribution, and compute X NL i with initial condition Y i. We also compute the covariance ki T, T of the linear correction. The acceptance probability for this update is α = N Y i X NL i, ki T, T N Zi :N pzi :N Zi :N N Y i X NL i, k i T, T N Z i :N pzi :N Zi :N 6 After updating the Gaussian inputs, we make a global update for the θ parameter. The acceptance probability for this move is n N Y i X NL i, ki α = T, T pθ pθ θ N Y i X NL i, k i T, T pθpθ θ, 7 i= where X NL i and k i T, T are computed using the proposed value of θ. We noted earlier that when j is large, Z j governs the small-time oscillations of the diffusion process. One should not expect to gain much information about the value of Z j when we have large interobservation times. We find this to be the case in our experiments - the posterior distribution of Z j:n approaches a spherical Gaussian distribution when j > 3. For this reason, we employ a Gaussian random walk proposal in Z with stepsize σ RW =.45, and proposals for Z :N are drawn independently from the standard normal distribution. 6

7 In the presence of observation noise, we proceed roughly as before. Recall that we make observations Y i = X i + ɛ i. We draw proposals Z i :N and ɛ i. The initial condition for XNL i is now Y i ɛ i. However, one must make an important modification to the algorithm. Suppose we propose an update of ˆX i and it is accepted. If we subsequently propose an update for ˆX i+ and it is rejected, then the initial condition for ˆX i+ will be inconsistent with the current state of the chain it will be Y i ɛ i instead of Y i ɛ i. For this reason, we must propose joint updates for ˆX i, ɛ i, ˆX i+. If the variance of the observation noise is high, it may be more efficient to target the joint posterior distribution p θ, {Z i :N, XL i } Y :n. 6 Numerical Experiments The double-well diffusion is a widely-used benchmark for nonlinear inference problems [4, 3, 33, 34]. It has been used to model systems that exhibit switching behaviour or bistability [, 35]. It possesses nonlinear features that are sufficient to demonstrate the shortcomings of some existing inference methods, and how our approach overcomes these issues. The dynamics of the process are given by dx t = αx t γ X t dt + BdWt. 8 The process X has a bimodal stationary distribution, with modes at x = ±γ. The parameter α governs the rate at which sample trajectories are pushed toward either mode. If B is small in comparison to α, mode-switching occurs relatively rarely. Figure b shows a trajectory of a double-well diffusion over units of time, with observations at times {,,..., }. We used the parameters α =, γ =, B =. The variance of the observation noise was set to Σ =.5. As we mentioned earlier, particle MCMC performs well in low-dimensional inference problems. For this reason, the results of a particle MCMC inference algorithm with N =, particles are used as ground truth. Our algorithm used N = 3 Gaussian inputs with a linear correction. We used the Fourier cosine series 8 as an orthonormal basis. We compare our Gibbs sampler to that of Golightly and Wilkinson [5], for which we use an Euler discretisation with stepsize t =.5. Each algorithm drew 7, samples from the posterior distribution, moving through the parameter space in a Gaussian random walk. We placed an exponential4 prior on γ and an exponential prior on α and B. For this particular choice of parameters, both Gibbs samplers give a good approximation to the true posterior. Figure shows histograms of the marginal posterior distributions of α, γ, B for each algorithm. pα α a pα Y : pγ γ b pγ Y : pb B c pb Y : Figure : Marginal posterior distributions for α, γ, B conditional on observed data. The solid black line is the output of a particle MCMC method, taken as ground truth. The broken red line is the output of the linear proposal method, and the broken and dotted blue line is the density estimate from the coloured noise expansion method. We see that both methods give a good approximation to the ground truth. Gibbs samplers that have been used in the past rely on making proposals by conditioning a linear diffusion to hit a target, and subsequently accepting or rejecting those proposals. Over short timescales, or for problems that are not highly nonlinear, this can be an effective strategy. However, as the timescale increases, the proposal and target become quite dissimilar see Figure a. 7

8 For our second experiment, we simulate a double well process with α, γ, B =,.5,. We make noisy observations with t i t i = 3 and Σ =.. The algorithms target the posterior distribution over γ, with α and B fixed at their true values. From our previous discussion, one might expect the linear proposal strategy to perform poorly in this more nonlinear setting. This is indeed the case. As in the previous experiment, we used a linear proposal Gibbs sampler with Euler stepsize dt =.5. In the path update stage, fewer than.% of proposals were accepted. On the other hand, the coloured noise expansion method used N = 7 Gaussian inputs with a linear correction and was able to approximate the posterior accurately. Figure 3 shows histograms of the results. Note the different scaling of the rightmost plot. pγ γ a Particle MCMC pγ γ b Coloured noise expansion method pγ γ c Linear proposal method Figure 3: pγ Y :, B, α after ten observations with a relatively large inter-observation time. We drew data from a double well process with α, γ, B =,.5,. The coloured noise expansion method matches the ground truth, whereas the linear proposal method is inconsistent with the data. 7 Discussion and Future Work We have seen that the standard linear proposal/correction strategy can fail for highly nonlinear problems. Our inference method avoids the linear correction step, instead targeting the posterior over input variables directly. With regard to computational efficiency, it is difficult to give an authoritative analysis because both our method and the linear proposal method are complex, with several parameters to tune. In our experiments, the algorithms terminated in a roughly similar length of time though no serious attempt was made to optimise the runtime of either method. With regard to our method, several questions remain open. The accuracy of our algorithm depends on the choice of basis functions {φ i }. At present, it is not clear how to make this choice optimally in the general setting. In the linear case, it is possible to show that one can achieve the accuracy of the Karhunen-Loeve decomposition, which is theoretically optimal. One can also set the error at a single time t to zero with a judicious choice of a single basis function. We aim to present these results in a paper that is currently under preparation. We used a Taylor expansion to compute the covariance of the correction term. However, it may be fruitful to use more sophisticated ideas, collectively known as statistical linearisation methods. In this paper, we restricted our attention to processes with a state-independent diffusion coefficient so that the covariance of the correction term could be computed. We may be able to extend this methodology to process with state-dependent noise - certainly one could achieve this by taking a -th order Taylor expansion about X NL. Whether it is possible to improve upon this idea is a matter for further investigation. Acknowledgments Simon Lyons was supported by Microsoft Research, Cambridge. References [] R.C. Merton. Theory of rational option pricing. The Bell Journal of Economics and Management Science, 4:4 83, 973. [] D.T. Gillespie. The chemical Langevin equation. Journal of Chemical Physics, 3,:97 36,. 8

9 [3] G. Kallianpur. Weak convergence of stochastic neuronal models. Stochastic Methods in Biology, 7:6 45, 987. [4] H.A. Dijkstra, L.M. Frankcombe, and A.S. von der Heydt. A stochastic dynamical systems view of the Atlantic Multidecadal Oscillation. Philosophical Transactions of the Royal Society A, 366: , 8. [5] L. Murray and A. Storkey. Continuous time particle filtering for fmri. Advances in Neural Information Processing Systems, :49 56, 8. [6] J. Daunizeau, K.J. Friston, and S.J. Kiebel. Variational Bayesian identification and prediction of stochastic nonlinear dynamic causal models. Physica D, pages 89 8, 9. [7] L.M. Murray and A.J. Storkey. Particle smoothing in continuous time: A fast approach via density estimation. IEEE Transactions on Signal Processing, 59:7 6,. [8] W. Feller. An Introduction to Probability Theory and its Applications, Volume II. Wiley, 97. [9] I. Karatzas and S.E. Shreve. Brownian Motion and Stochastic Calculus. Springer, 99. [] B. Oksendal. Stochastic Differential Equations. Springer, 7. [] C.W. Gardiner. Handbook of Stochastic Methods for Physics, Chemistry and the Natural Sciences. Springer-Verlag, 983. [] Y. Aït-Sahalia. Closed-form likelihood expansions for multivariate diffusions. The Annals of Statistics, 36:96 937, 8. [3] A. Beskos, O. Papaspiliopoulos, and G.O. Roberts. Monte-Carlo maximum likelihood estimation for discretely observed diffusion processes. Annals of Statistics, 37:3 45, 9. [4] A. Beskos, O. Papaspiliopoulos, G.O. Roberts, and P. Fearnhead. Exact and computationally efficient likelihood-based estimation for discretely observed diffusion processes with discussion. Journal of the Royal Statistical Society: Series B Statistical Methodology, 68:333 38, 6. [5] A. Golightly and D.J. Wilkinson. Bayesian inference for nonlinear multivariate diffusion models observed with error. CSDA, 5: , 8. [6] S. Chib, M.K. Pitt, and N. Shepard. Likelihood-based inference for diffusion models. Working Paper, 4. [7] G.B. Durham and A.R. Gallant. Numerical techniques for maximum likelihood estimation of continuoustime diffusion processes with comments. Journal of Business and Economic Statistics, :97 338,. [8] A. Beskos, O. Papaspiliopoulos, and G.O. Roberts. Retrospective exact simulation of diffusion sample paths with applications. Bernoulli, 6:77, 6. [9] D. Rimmer, A. Doucet, and W.J. Fitzgerald. Particle filters for stochastic differential equations of nonlinear diffusions. Technical report, Cambridge University Engineering Department, 5. [] Christophe Andrieu, Arnaud Doucet, and Roman Holenstein. Particle Markov Chain Monte Carlo methods. Journal of the Royal Statistical Society, 7: 33,. [] C. Snyder, T. Bengtsson, P. Bickel, and J. Anderson. Obstacles to high-dimensional particle filtering. Monthly Weather Review, 36: , 8. [] Y. Aït-Sahalia. Maximum likelihood estimation of discretely sampled diffusions: a closed-form approximation approach. Econometrica, 7:3 6,. [3] C. Archambeau, D. Cornford, M. Opper, and J. Shawe-Taylor. Gaussian process approximations of stochastic differential equations. JMLR: Workshop and Conference Proceedings, : 6, 7. [4] C. Archambeau, M. Opper, Y. Shen, D. Cornford, and J. Shawe-Taylor. Variational inference for diffusion processes. In Advances in Neural Information Processing Systems NIPS 7, 8. [5] S. Särkkä. On unscented Kalman filtering for state estimation of continuous-time nonlinear systems. IEEE Transactions on Automatic Control, 5:63 64, 7. [6] A.H. Jazwinski. Stochastic processes and filtering theory, volume 63. Academic Pr, 97. [7] H. Singer. Nonlinear continuous time modeling approaches in panel research. Statistica Neerlandica, 6:9 57, 8. [8] S. Corlay and P. Gilles. Functional quantization based stratified sampling methods. Arxiv preprint Arxiv:8.444,. [9] W. Luo. Wiener chaos expansion and numerical solutions of stochastic partial differential equations. PhD thesis, California Institute of Technology, 6. [3] P.E. Kloeden and E. Platen. Numerical Solution of Stochastic Differential Equations. Springer, 999. [3] A.F. Bastani and S.M. Hosseini. A new adaptive Runge-Kutta method for stochastic differential equations. Journal of Computational and Applied Mathematics, 6:63 644, 7. [3] Y. Shen, C. Archambeau, D. Cornford, M. Opper, J. Shawe-Taylor, and R. Barillec. A comparison of variational and Markov chain Monte Carlo methods for inference in partially observed stochastic dynamic systems. Journal of Signal Processing Systems, 6:5 59,. [33] H. Singer. Parameter estimation of nonlinear stochastic differential equations: simulated maximum likelihood versus extended Kalman filter and Itô-Taylor expansion. Journal of Computational and Graphical Statistics, 4:97 995,. [34] M. Opper, A. Ruttor, and G. Sanguinetti. Approximate inference in continuous time Gaussian-jump processes. Advances in Neural Information Processing Systems, 3:83 839,. [35] N.G. van Kampen. Stochastic processes in physics and chemistry. North holland, 7. 9

A variational radial basis function approximation for diffusion processes

A variational radial basis function approximation for diffusion processes A variational radial basis function approximation for diffusion processes Michail D. Vrettas, Dan Cornford and Yuan Shen Aston University - Neural Computing Research Group Aston Triangle, Birmingham B4

More information

Gaussian Process Approximations of Stochastic Differential Equations

Gaussian Process Approximations of Stochastic Differential Equations Gaussian Process Approximations of Stochastic Differential Equations Cédric Archambeau Centre for Computational Statistics and Machine Learning University College London c.archambeau@cs.ucl.ac.uk CSML

More information

Nonparametric Drift Estimation for Stochastic Differential Equations

Nonparametric Drift Estimation for Stochastic Differential Equations Nonparametric Drift Estimation for Stochastic Differential Equations Gareth Roberts 1 Department of Statistics University of Warwick Brazilian Bayesian meeting, March 2010 Joint work with O. Papaspiliopoulos,

More information

Computational statistics

Computational statistics Computational statistics Markov Chain Monte Carlo methods Thierry Denœux March 2017 Thierry Denœux Computational statistics March 2017 1 / 71 Contents of this chapter When a target density f can be evaluated

More information

Sequential Monte Carlo Methods for Bayesian Computation

Sequential Monte Carlo Methods for Bayesian Computation Sequential Monte Carlo Methods for Bayesian Computation A. Doucet Kyoto Sept. 2012 A. Doucet (MLSS Sept. 2012) Sept. 2012 1 / 136 Motivating Example 1: Generic Bayesian Model Let X be a vector parameter

More information

Bayesian Inference for Discretely Sampled Diffusion Processes: A New MCMC Based Approach to Inference

Bayesian Inference for Discretely Sampled Diffusion Processes: A New MCMC Based Approach to Inference Bayesian Inference for Discretely Sampled Diffusion Processes: A New MCMC Based Approach to Inference Osnat Stramer 1 and Matthew Bognar 1 Department of Statistics and Actuarial Science, University of

More information

Handbook of Stochastic Methods

Handbook of Stochastic Methods C. W. Gardiner Handbook of Stochastic Methods for Physics, Chemistry and the Natural Sciences Third Edition With 30 Figures Springer Contents 1. A Historical Introduction 1 1.1 Motivation I 1.2 Some Historical

More information

CPSC 540: Machine Learning

CPSC 540: Machine Learning CPSC 540: Machine Learning MCMC and Non-Parametric Bayes Mark Schmidt University of British Columbia Winter 2016 Admin I went through project proposals: Some of you got a message on Piazza. No news is

More information

Lecture 6: Bayesian Inference in SDE Models

Lecture 6: Bayesian Inference in SDE Models Lecture 6: Bayesian Inference in SDE Models Bayesian Filtering and Smoothing Point of View Simo Särkkä Aalto University Simo Särkkä (Aalto) Lecture 6: Bayesian Inference in SDEs 1 / 45 Contents 1 SDEs

More information

STA 4273H: Statistical Machine Learning

STA 4273H: Statistical Machine Learning STA 4273H: Statistical Machine Learning Russ Salakhutdinov Department of Computer Science! Department of Statistical Sciences! rsalakhu@cs.toronto.edu! h0p://www.cs.utoronto.ca/~rsalakhu/ Lecture 7 Approximate

More information

16 : Markov Chain Monte Carlo (MCMC)

16 : Markov Chain Monte Carlo (MCMC) 10-708: Probabilistic Graphical Models 10-708, Spring 2014 16 : Markov Chain Monte Carlo MCMC Lecturer: Matthew Gormley Scribes: Yining Wang, Renato Negrinho 1 Sampling from low-dimensional distributions

More information

Handbook of Stochastic Methods

Handbook of Stochastic Methods Springer Series in Synergetics 13 Handbook of Stochastic Methods for Physics, Chemistry and the Natural Sciences von Crispin W Gardiner Neuausgabe Handbook of Stochastic Methods Gardiner schnell und portofrei

More information

Gaussian Process Approximations of Stochastic Differential Equations

Gaussian Process Approximations of Stochastic Differential Equations Gaussian Process Approximations of Stochastic Differential Equations Cédric Archambeau Dan Cawford Manfred Opper John Shawe-Taylor May, 2006 1 Introduction Some of the most complex models routinely run

More information

Labor-Supply Shifts and Economic Fluctuations. Technical Appendix

Labor-Supply Shifts and Economic Fluctuations. Technical Appendix Labor-Supply Shifts and Economic Fluctuations Technical Appendix Yongsung Chang Department of Economics University of Pennsylvania Frank Schorfheide Department of Economics University of Pennsylvania January

More information

A Note on Auxiliary Particle Filters

A Note on Auxiliary Particle Filters A Note on Auxiliary Particle Filters Adam M. Johansen a,, Arnaud Doucet b a Department of Mathematics, University of Bristol, UK b Departments of Statistics & Computer Science, University of British Columbia,

More information

An introduction to Sequential Monte Carlo

An introduction to Sequential Monte Carlo An introduction to Sequential Monte Carlo Thang Bui Jes Frellsen Department of Engineering University of Cambridge Research and Communication Club 6 February 2014 1 Sequential Monte Carlo (SMC) methods

More information

Exact Simulation of Diffusions and Jump Diffusions

Exact Simulation of Diffusions and Jump Diffusions Exact Simulation of Diffusions and Jump Diffusions A work by: Prof. Gareth O. Roberts Dr. Alexandros Beskos Dr. Omiros Papaspiliopoulos Dr. Bruno Casella 28 th May, 2008 Content 1 Exact Algorithm Construction

More information

Bayesian Methods for Machine Learning

Bayesian Methods for Machine Learning Bayesian Methods for Machine Learning CS 584: Big Data Analytics Material adapted from Radford Neal s tutorial (http://ftp.cs.utoronto.ca/pub/radford/bayes-tut.pdf), Zoubin Ghahramni (http://hunch.net/~coms-4771/zoubin_ghahramani_bayesian_learning.pdf),

More information

Exercises Tutorial at ICASSP 2016 Learning Nonlinear Dynamical Models Using Particle Filters

Exercises Tutorial at ICASSP 2016 Learning Nonlinear Dynamical Models Using Particle Filters Exercises Tutorial at ICASSP 216 Learning Nonlinear Dynamical Models Using Particle Filters Andreas Svensson, Johan Dahlin and Thomas B. Schön March 18, 216 Good luck! 1 [Bootstrap particle filter for

More information

Lecture 4: Numerical Solution of SDEs, Itô Taylor Series, Gaussian Process Approximations

Lecture 4: Numerical Solution of SDEs, Itô Taylor Series, Gaussian Process Approximations Lecture 4: Numerical Solution of SDEs, Itô Taylor Series, Gaussian Process Approximations Simo Särkkä Aalto University Tampere University of Technology Lappeenranta University of Technology Finland November

More information

Lecture 4: Numerical Solution of SDEs, Itô Taylor Series, Gaussian Approximations

Lecture 4: Numerical Solution of SDEs, Itô Taylor Series, Gaussian Approximations Lecture 4: Numerical Solution of SDEs, Itô Taylor Series, Gaussian Approximations Simo Särkkä Aalto University, Finland November 18, 2014 Simo Särkkä (Aalto) Lecture 4: Numerical Solution of SDEs November

More information

Lecture 2: From Linear Regression to Kalman Filter and Beyond

Lecture 2: From Linear Regression to Kalman Filter and Beyond Lecture 2: From Linear Regression to Kalman Filter and Beyond Department of Biomedical Engineering and Computational Science Aalto University January 26, 2012 Contents 1 Batch and Recursive Estimation

More information

Numerical Integration of SDEs: A Short Tutorial

Numerical Integration of SDEs: A Short Tutorial Numerical Integration of SDEs: A Short Tutorial Thomas Schaffter January 19, 010 1 Introduction 1.1 Itô and Stratonovich SDEs 1-dimensional stochastic differentiable equation (SDE) is given by [6, 7] dx

More information

Kernel Sequential Monte Carlo

Kernel Sequential Monte Carlo Kernel Sequential Monte Carlo Ingmar Schuster (Paris Dauphine) Heiko Strathmann (University College London) Brooks Paige (Oxford) Dino Sejdinovic (Oxford) * equal contribution April 25, 2016 1 / 37 Section

More information

17 : Markov Chain Monte Carlo

17 : Markov Chain Monte Carlo 10-708: Probabilistic Graphical Models, Spring 2015 17 : Markov Chain Monte Carlo Lecturer: Eric P. Xing Scribes: Heran Lin, Bin Deng, Yun Huang 1 Review of Monte Carlo Methods 1.1 Overview Monte Carlo

More information

Towards inference for skewed alpha stable Levy processes

Towards inference for skewed alpha stable Levy processes Towards inference for skewed alpha stable Levy processes Simon Godsill and Tatjana Lemke Signal Processing and Communications Lab. University of Cambridge www-sigproc.eng.cam.ac.uk/~sjg Overview Motivation

More information

April 20th, Advanced Topics in Machine Learning California Institute of Technology. Markov Chain Monte Carlo for Machine Learning

April 20th, Advanced Topics in Machine Learning California Institute of Technology. Markov Chain Monte Carlo for Machine Learning for for Advanced Topics in California Institute of Technology April 20th, 2017 1 / 50 Table of Contents for 1 2 3 4 2 / 50 History of methods for Enrico Fermi used to calculate incredibly accurate predictions

More information

Markov chain Monte Carlo algorithms for SDE parameter estimation

Markov chain Monte Carlo algorithms for SDE parameter estimation Markov chain Monte Carlo algorithms for SDE parameter estimation Andrew Golightly and Darren J. Wilkinson Abstract This chapter considers stochastic differential equations for Systems Biology models derived

More information

Dynamic System Identification using HDMR-Bayesian Technique

Dynamic System Identification using HDMR-Bayesian Technique Dynamic System Identification using HDMR-Bayesian Technique *Shereena O A 1) and Dr. B N Rao 2) 1), 2) Department of Civil Engineering, IIT Madras, Chennai 600036, Tamil Nadu, India 1) ce14d020@smail.iitm.ac.in

More information

Stochastic differential equation models in biology Susanne Ditlevsen

Stochastic differential equation models in biology Susanne Ditlevsen Stochastic differential equation models in biology Susanne Ditlevsen Introduction This chapter is concerned with continuous time processes, which are often modeled as a system of ordinary differential

More information

Kernel adaptive Sequential Monte Carlo

Kernel adaptive Sequential Monte Carlo Kernel adaptive Sequential Monte Carlo Ingmar Schuster (Paris Dauphine) Heiko Strathmann (University College London) Brooks Paige (Oxford) Dino Sejdinovic (Oxford) December 7, 2015 1 / 36 Section 1 Outline

More information

Sequential Monte Carlo Methods in High Dimensions

Sequential Monte Carlo Methods in High Dimensions Sequential Monte Carlo Methods in High Dimensions Alexandros Beskos Statistical Science, UCL Oxford, 24th September 2012 Joint work with: Dan Crisan, Ajay Jasra, Nik Kantas, Andrew Stuart Imperial College,

More information

Introduction to Machine Learning CMU-10701

Introduction to Machine Learning CMU-10701 Introduction to Machine Learning CMU-10701 Markov Chain Monte Carlo Methods Barnabás Póczos & Aarti Singh Contents Markov Chain Monte Carlo Methods Goal & Motivation Sampling Rejection Importance Markov

More information

Monte Carlo Methods. Leon Gu CSD, CMU

Monte Carlo Methods. Leon Gu CSD, CMU Monte Carlo Methods Leon Gu CSD, CMU Approximate Inference EM: y-observed variables; x-hidden variables; θ-parameters; E-step: q(x) = p(x y, θ t 1 ) M-step: θ t = arg max E q(x) [log p(y, x θ)] θ Monte

More information

Prediction of ESTSP Competition Time Series by Unscented Kalman Filter and RTS Smoother

Prediction of ESTSP Competition Time Series by Unscented Kalman Filter and RTS Smoother Prediction of ESTSP Competition Time Series by Unscented Kalman Filter and RTS Smoother Simo Särkkä, Aki Vehtari and Jouko Lampinen Helsinki University of Technology Department of Electrical and Communications

More information

Learning Static Parameters in Stochastic Processes

Learning Static Parameters in Stochastic Processes Learning Static Parameters in Stochastic Processes Bharath Ramsundar December 14, 2012 1 Introduction Consider a Markovian stochastic process X T evolving (perhaps nonlinearly) over time variable T. We

More information

Inferring biological dynamics Iterated filtering (IF)

Inferring biological dynamics Iterated filtering (IF) Inferring biological dynamics 101 3. Iterated filtering (IF) IF originated in 2006 [6]. For plug-and-play likelihood-based inference on POMP models, there are not many alternatives. Directly estimating

More information

Hastings-within-Gibbs Algorithm: Introduction and Application on Hierarchical Model

Hastings-within-Gibbs Algorithm: Introduction and Application on Hierarchical Model UNIVERSITY OF TEXAS AT SAN ANTONIO Hastings-within-Gibbs Algorithm: Introduction and Application on Hierarchical Model Liang Jing April 2010 1 1 ABSTRACT In this paper, common MCMC algorithms are introduced

More information

Nonparametric Bayesian Methods - Lecture I

Nonparametric Bayesian Methods - Lecture I Nonparametric Bayesian Methods - Lecture I Harry van Zanten Korteweg-de Vries Institute for Mathematics CRiSM Masterclass, April 4-6, 2016 Overview of the lectures I Intro to nonparametric Bayesian statistics

More information

Controlled sequential Monte Carlo

Controlled sequential Monte Carlo Controlled sequential Monte Carlo Jeremy Heng, Department of Statistics, Harvard University Joint work with Adrian Bishop (UTS, CSIRO), George Deligiannidis & Arnaud Doucet (Oxford) Bayesian Computation

More information

The Metropolis-Hastings Algorithm. June 8, 2012

The Metropolis-Hastings Algorithm. June 8, 2012 The Metropolis-Hastings Algorithm June 8, 22 The Plan. Understand what a simulated distribution is 2. Understand why the Metropolis-Hastings algorithm works 3. Learn how to apply the Metropolis-Hastings

More information

Lecture 4: Introduction to stochastic processes and stochastic calculus

Lecture 4: Introduction to stochastic processes and stochastic calculus Lecture 4: Introduction to stochastic processes and stochastic calculus Cédric Archambeau Centre for Computational Statistics and Machine Learning Department of Computer Science University College London

More information

M4A42 APPLIED STOCHASTIC PROCESSES

M4A42 APPLIED STOCHASTIC PROCESSES M4A42 APPLIED STOCHASTIC PROCESSES G.A. Pavliotis Department of Mathematics Imperial College London, UK LECTURE 1 12/10/2009 Lectures: Mondays 09:00-11:00, Huxley 139, Tuesdays 09:00-10:00, Huxley 144.

More information

MCMC for big data. Geir Storvik. BigInsight lunch - May Geir Storvik MCMC for big data BigInsight lunch - May / 17

MCMC for big data. Geir Storvik. BigInsight lunch - May Geir Storvik MCMC for big data BigInsight lunch - May / 17 MCMC for big data Geir Storvik BigInsight lunch - May 2 2018 Geir Storvik MCMC for big data BigInsight lunch - May 2 2018 1 / 17 Outline Why ordinary MCMC is not scalable Different approaches for making

More information

Imprecise Filtering for Spacecraft Navigation

Imprecise Filtering for Spacecraft Navigation Imprecise Filtering for Spacecraft Navigation Tathagata Basu Cristian Greco Thomas Krak Durham University Strathclyde University Ghent University Filtering for Spacecraft Navigation The General Problem

More information

State-Space Methods for Inferring Spike Trains from Calcium Imaging

State-Space Methods for Inferring Spike Trains from Calcium Imaging State-Space Methods for Inferring Spike Trains from Calcium Imaging Joshua Vogelstein Johns Hopkins April 23, 2009 Joshua Vogelstein (Johns Hopkins) State-Space Calcium Imaging April 23, 2009 1 / 78 Outline

More information

Simulation Method for Solving Stochastic Differential Equations with Constant Diffusion Coefficients

Simulation Method for Solving Stochastic Differential Equations with Constant Diffusion Coefficients Journal of mathematics and computer Science 8 (2014) 28-32 Simulation Method for Solving Stochastic Differential Equations with Constant Diffusion Coefficients Behrouz Fathi Vajargah Department of statistics,

More information

Calibration of Stochastic Volatility Models using Particle Markov Chain Monte Carlo Methods

Calibration of Stochastic Volatility Models using Particle Markov Chain Monte Carlo Methods Calibration of Stochastic Volatility Models using Particle Markov Chain Monte Carlo Methods Jonas Hallgren 1 1 Department of Mathematics KTH Royal Institute of Technology Stockholm, Sweden BFS 2012 June

More information

Lecture on Parameter Estimation for Stochastic Differential Equations. Erik Lindström

Lecture on Parameter Estimation for Stochastic Differential Equations. Erik Lindström Lecture on Parameter Estimation for Stochastic Differential Equations Erik Lindström Recap We are interested in the parameters θ in the Stochastic Integral Equations X(t) = X(0) + t 0 µ θ (s, X(s))ds +

More information

Markov Chain Monte Carlo

Markov Chain Monte Carlo 1 Motivation 1.1 Bayesian Learning Markov Chain Monte Carlo Yale Chang In Bayesian learning, given data X, we make assumptions on the generative process of X by introducing hidden variables Z: p(z): prior

More information

Monte Carlo Methods. Handbook of. University ofqueensland. Thomas Taimre. Zdravko I. Botev. Dirk P. Kroese. Universite de Montreal

Monte Carlo Methods. Handbook of. University ofqueensland. Thomas Taimre. Zdravko I. Botev. Dirk P. Kroese. Universite de Montreal Handbook of Monte Carlo Methods Dirk P. Kroese University ofqueensland Thomas Taimre University ofqueensland Zdravko I. Botev Universite de Montreal A JOHN WILEY & SONS, INC., PUBLICATION Preface Acknowledgments

More information

A Search and Jump Algorithm for Markov Chain Monte Carlo Sampling. Christopher Jennison. Adriana Ibrahim. Seminar at University of Kuwait

A Search and Jump Algorithm for Markov Chain Monte Carlo Sampling. Christopher Jennison. Adriana Ibrahim. Seminar at University of Kuwait A Search and Jump Algorithm for Markov Chain Monte Carlo Sampling Christopher Jennison Department of Mathematical Sciences, University of Bath, UK http://people.bath.ac.uk/mascj Adriana Ibrahim Institute

More information

Probabilistic Graphical Models Lecture 17: Markov chain Monte Carlo

Probabilistic Graphical Models Lecture 17: Markov chain Monte Carlo Probabilistic Graphical Models Lecture 17: Markov chain Monte Carlo Andrew Gordon Wilson www.cs.cmu.edu/~andrewgw Carnegie Mellon University March 18, 2015 1 / 45 Resources and Attribution Image credits,

More information

List of projects. FMS020F NAMS002 Statistical inference for partially observed stochastic processes, 2016

List of projects. FMS020F NAMS002 Statistical inference for partially observed stochastic processes, 2016 List of projects FMS020F NAMS002 Statistical inference for partially observed stochastic processes, 206 Work in groups of two (if this is absolutely not possible for some reason, please let the lecturers

More information

RESEARCH ARTICLE. Online quantization in nonlinear filtering

RESEARCH ARTICLE. Online quantization in nonlinear filtering Journal of Statistical Computation & Simulation Vol. 00, No. 00, Month 200x, 3 RESEARCH ARTICLE Online quantization in nonlinear filtering A. Feuer and G. C. Goodwin Received 00 Month 200x; in final form

More information

Lagrangian Data Assimilation and Manifold Detection for a Point-Vortex Model. David Darmon, AMSC Kayo Ide, AOSC, IPST, CSCAMM, ESSIC

Lagrangian Data Assimilation and Manifold Detection for a Point-Vortex Model. David Darmon, AMSC Kayo Ide, AOSC, IPST, CSCAMM, ESSIC Lagrangian Data Assimilation and Manifold Detection for a Point-Vortex Model David Darmon, AMSC Kayo Ide, AOSC, IPST, CSCAMM, ESSIC Background Data Assimilation Iterative process Forecast Analysis Background

More information

Advanced Computational Methods in Statistics: Lecture 5 Sequential Monte Carlo/Particle Filtering

Advanced Computational Methods in Statistics: Lecture 5 Sequential Monte Carlo/Particle Filtering Advanced Computational Methods in Statistics: Lecture 5 Sequential Monte Carlo/Particle Filtering Axel Gandy Department of Mathematics Imperial College London http://www2.imperial.ac.uk/~agandy London

More information

Sequential Monte Carlo Samplers for Applications in High Dimensions

Sequential Monte Carlo Samplers for Applications in High Dimensions Sequential Monte Carlo Samplers for Applications in High Dimensions Alexandros Beskos National University of Singapore KAUST, 26th February 2014 Joint work with: Dan Crisan, Ajay Jasra, Nik Kantas, Alex

More information

18 : Advanced topics in MCMC. 1 Gibbs Sampling (Continued from the last lecture)

18 : Advanced topics in MCMC. 1 Gibbs Sampling (Continued from the last lecture) 10-708: Probabilistic Graphical Models 10-708, Spring 2014 18 : Advanced topics in MCMC Lecturer: Eric P. Xing Scribes: Jessica Chemali, Seungwhan Moon 1 Gibbs Sampling (Continued from the last lecture)

More information

Modeling Ultra-High-Frequency Multivariate Financial Data by Monte Carlo Simulation Methods

Modeling Ultra-High-Frequency Multivariate Financial Data by Monte Carlo Simulation Methods Outline Modeling Ultra-High-Frequency Multivariate Financial Data by Monte Carlo Simulation Methods Ph.D. Student: Supervisor: Marco Minozzo Dipartimento di Scienze Economiche Università degli Studi di

More information

Bayesian parameter inference for stochastic biochemical network models using particle MCMC

Bayesian parameter inference for stochastic biochemical network models using particle MCMC Bayesian parameter inference for stochastic biochemical network models using particle MCMC Andrew Golightly Darren J. Wilkinson May 16, 2011 Abstract Computational systems biology is concerned with the

More information

MCMC algorithms for fitting Bayesian models

MCMC algorithms for fitting Bayesian models MCMC algorithms for fitting Bayesian models p. 1/1 MCMC algorithms for fitting Bayesian models Sudipto Banerjee sudiptob@biostat.umn.edu University of Minnesota MCMC algorithms for fitting Bayesian models

More information

Stochastic Integration (Simple Version)

Stochastic Integration (Simple Version) Stochastic Integration (Simple Version) Tuesday, March 17, 2015 2:03 PM Reading: Gardiner Secs. 4.1-4.3, 4.4.4 But there are boundary issues when (if ) so we can't apply the standard delta function integration

More information

Bayesian Inference of Stochastic Pursuit Models from Basketball Tracking Data

Bayesian Inference of Stochastic Pursuit Models from Basketball Tracking Data Bayesian Inference of Stochastic Pursuit Models from Basketball Tracking Data Harish S. Bhat, R. W. M. A. Madushani, and Shagun Rawat 1 Introduction In 2010, the National Basketball Association (NBA) began

More information

Ergodicity in data assimilation methods

Ergodicity in data assimilation methods Ergodicity in data assimilation methods David Kelly Andy Majda Xin Tong Courant Institute New York University New York NY www.dtbkelly.com April 15, 2016 ETH Zurich David Kelly (CIMS) Data assimilation

More information

Bayesian Calibration of Simulators with Structured Discretization Uncertainty

Bayesian Calibration of Simulators with Structured Discretization Uncertainty Bayesian Calibration of Simulators with Structured Discretization Uncertainty Oksana A. Chkrebtii Department of Statistics, The Ohio State University Joint work with Matthew T. Pratola (Statistics, The

More information

Statistical Inference and Methods

Statistical Inference and Methods Department of Mathematics Imperial College London d.stephens@imperial.ac.uk http://stats.ma.ic.ac.uk/ das01/ 31st January 2006 Part VI Session 6: Filtering and Time to Event Data Session 6: Filtering and

More information

Inference in state-space models with multiple paths from conditional SMC

Inference in state-space models with multiple paths from conditional SMC Inference in state-space models with multiple paths from conditional SMC Sinan Yıldırım (Sabancı) joint work with Christophe Andrieu (Bristol), Arnaud Doucet (Oxford) and Nicolas Chopin (ENSAE) September

More information

Introduction to Machine Learning

Introduction to Machine Learning Introduction to Machine Learning Brown University CSCI 1950-F, Spring 2012 Prof. Erik Sudderth Lecture 25: Markov Chain Monte Carlo (MCMC) Course Review and Advanced Topics Many figures courtesy Kevin

More information

Introduction to Particle Filters for Data Assimilation

Introduction to Particle Filters for Data Assimilation Introduction to Particle Filters for Data Assimilation Mike Dowd Dept of Mathematics & Statistics (and Dept of Oceanography Dalhousie University, Halifax, Canada STATMOS Summer School in Data Assimila5on,

More information

Sequential Monte Carlo Methods (for DSGE Models)

Sequential Monte Carlo Methods (for DSGE Models) Sequential Monte Carlo Methods (for DSGE Models) Frank Schorfheide University of Pennsylvania, PIER, CEPR, and NBER October 23, 2017 Some References These lectures use material from our joint work: Tempered

More information

Bayesian room-acoustic modal analysis

Bayesian room-acoustic modal analysis Bayesian room-acoustic modal analysis Wesley Henderson a) Jonathan Botts b) Ning Xiang c) Graduate Program in Architectural Acoustics, School of Architecture, Rensselaer Polytechnic Institute, Troy, New

More information

A nonparametric method of multi-step ahead forecasting in diffusion processes

A nonparametric method of multi-step ahead forecasting in diffusion processes A nonparametric method of multi-step ahead forecasting in diffusion processes Mariko Yamamura a, Isao Shoji b a School of Pharmacy, Kitasato University, Minato-ku, Tokyo, 108-8641, Japan. b Graduate School

More information

A note on Reversible Jump Markov Chain Monte Carlo

A note on Reversible Jump Markov Chain Monte Carlo A note on Reversible Jump Markov Chain Monte Carlo Hedibert Freitas Lopes Graduate School of Business The University of Chicago 5807 South Woodlawn Avenue Chicago, Illinois 60637 February, 1st 2006 1 Introduction

More information

On the Optimal Scaling of the Modified Metropolis-Hastings algorithm

On the Optimal Scaling of the Modified Metropolis-Hastings algorithm On the Optimal Scaling of the Modified Metropolis-Hastings algorithm K. M. Zuev & J. L. Beck Division of Engineering and Applied Science California Institute of Technology, MC 4-44, Pasadena, CA 925, USA

More information

CS242: Probabilistic Graphical Models Lecture 7B: Markov Chain Monte Carlo & Gibbs Sampling

CS242: Probabilistic Graphical Models Lecture 7B: Markov Chain Monte Carlo & Gibbs Sampling CS242: Probabilistic Graphical Models Lecture 7B: Markov Chain Monte Carlo & Gibbs Sampling Professor Erik Sudderth Brown University Computer Science October 27, 2016 Some figures and materials courtesy

More information

Computer Vision Group Prof. Daniel Cremers. 11. Sampling Methods

Computer Vision Group Prof. Daniel Cremers. 11. Sampling Methods Prof. Daniel Cremers 11. Sampling Methods Sampling Methods Sampling Methods are widely used in Computer Science as an approximation of a deterministic algorithm to represent uncertainty without a parametric

More information

Accurate approximation of stochastic differential equations

Accurate approximation of stochastic differential equations Accurate approximation of stochastic differential equations Simon J.A. Malham and Anke Wiese (Heriot Watt University, Edinburgh) Birmingham: 6th February 29 Stochastic differential equations dy t = V (y

More information

MARKOV CHAIN MONTE CARLO

MARKOV CHAIN MONTE CARLO MARKOV CHAIN MONTE CARLO RYAN WANG Abstract. This paper gives a brief introduction to Markov Chain Monte Carlo methods, which offer a general framework for calculating difficult integrals. We start with

More information

Pseudo-marginal MCMC methods for inference in latent variable models

Pseudo-marginal MCMC methods for inference in latent variable models Pseudo-marginal MCMC methods for inference in latent variable models Arnaud Doucet Department of Statistics, Oxford University Joint work with George Deligiannidis (Oxford) & Mike Pitt (Kings) MCQMC, 19/08/2016

More information

Markov Chain Monte Carlo methods

Markov Chain Monte Carlo methods Markov Chain Monte Carlo methods Tomas McKelvey and Lennart Svensson Signal Processing Group Department of Signals and Systems Chalmers University of Technology, Sweden November 26, 2012 Today s learning

More information

Bagging During Markov Chain Monte Carlo for Smoother Predictions

Bagging During Markov Chain Monte Carlo for Smoother Predictions Bagging During Markov Chain Monte Carlo for Smoother Predictions Herbert K. H. Lee University of California, Santa Cruz Abstract: Making good predictions from noisy data is a challenging problem. Methods

More information

Brief introduction to Markov Chain Monte Carlo

Brief introduction to Markov Chain Monte Carlo Brief introduction to Department of Probability and Mathematical Statistics seminar Stochastic modeling in economics and finance November 7, 2011 Brief introduction to Content 1 and motivation Classical

More information

Particle Filters: Convergence Results and High Dimensions

Particle Filters: Convergence Results and High Dimensions Particle Filters: Convergence Results and High Dimensions Mark Coates mark.coates@mcgill.ca McGill University Department of Electrical and Computer Engineering Montreal, Quebec, Canada Bellairs 2012 Outline

More information

WEAK VERSIONS OF STOCHASTIC ADAMS-BASHFORTH AND SEMI-IMPLICIT LEAPFROG SCHEMES FOR SDES. 1. Introduction

WEAK VERSIONS OF STOCHASTIC ADAMS-BASHFORTH AND SEMI-IMPLICIT LEAPFROG SCHEMES FOR SDES. 1. Introduction WEAK VERSIONS OF STOCHASTIC ADAMS-BASHFORTH AND SEMI-IMPLICIT LEAPFROG SCHEMES FOR SDES BRIAN D. EWALD 1 Abstract. We consider the weak analogues of certain strong stochastic numerical schemes considered

More information

Bayesian Estimation of Input Output Tables for Russia

Bayesian Estimation of Input Output Tables for Russia Bayesian Estimation of Input Output Tables for Russia Oleg Lugovoy (EDF, RANE) Andrey Polbin (RANE) Vladimir Potashnikov (RANE) WIOD Conference April 24, 2012 Groningen Outline Motivation Objectives Bayesian

More information

Markov Chain Monte Carlo methods

Markov Chain Monte Carlo methods Markov Chain Monte Carlo methods By Oleg Makhnin 1 Introduction a b c M = d e f g h i 0 f(x)dx 1.1 Motivation 1.1.1 Just here Supresses numbering 1.1.2 After this 1.2 Literature 2 Method 2.1 New math As

More information

Lecture 2: From Linear Regression to Kalman Filter and Beyond

Lecture 2: From Linear Regression to Kalman Filter and Beyond Lecture 2: From Linear Regression to Kalman Filter and Beyond January 18, 2017 Contents 1 Batch and Recursive Estimation 2 Towards Bayesian Filtering 3 Kalman Filter and Bayesian Filtering and Smoothing

More information

ELEC633: Graphical Models

ELEC633: Graphical Models ELEC633: Graphical Models Tahira isa Saleem Scribe from 7 October 2008 References: Casella and George Exploring the Gibbs sampler (1992) Chib and Greenberg Understanding the Metropolis-Hastings algorithm

More information

27 : Distributed Monte Carlo Markov Chain. 1 Recap of MCMC and Naive Parallel Gibbs Sampling

27 : Distributed Monte Carlo Markov Chain. 1 Recap of MCMC and Naive Parallel Gibbs Sampling 10-708: Probabilistic Graphical Models 10-708, Spring 2014 27 : Distributed Monte Carlo Markov Chain Lecturer: Eric P. Xing Scribes: Pengtao Xie, Khoa Luu In this scribe, we are going to review the Parallel

More information

A Review of Pseudo-Marginal Markov Chain Monte Carlo

A Review of Pseudo-Marginal Markov Chain Monte Carlo A Review of Pseudo-Marginal Markov Chain Monte Carlo Discussed by: Yizhe Zhang October 21, 2016 Outline 1 Overview 2 Paper review 3 experiment 4 conclusion Motivation & overview Notation: θ denotes the

More information

STA 4273H: Statistical Machine Learning

STA 4273H: Statistical Machine Learning STA 4273H: Statistical Machine Learning Russ Salakhutdinov Department of Statistics! rsalakhu@utstat.toronto.edu! http://www.utstat.utoronto.ca/~rsalakhu/ Sidney Smith Hall, Room 6002 Lecture 3 Linear

More information

Thermostatic Controls for Noisy Gradient Systems and Applications to Machine Learning

Thermostatic Controls for Noisy Gradient Systems and Applications to Machine Learning Thermostatic Controls for Noisy Gradient Systems and Applications to Machine Learning Ben Leimkuhler University of Edinburgh Joint work with C. Matthews (Chicago), G. Stoltz (ENPC-Paris), M. Tretyakov

More information

Monte Carlo Approximation of Monte Carlo Filters

Monte Carlo Approximation of Monte Carlo Filters Monte Carlo Approximation of Monte Carlo Filters Adam M. Johansen et al. Collaborators Include: Arnaud Doucet, Axel Finke, Anthony Lee, Nick Whiteley 7th January 2014 Context & Outline Filtering in State-Space

More information

Discretization of SDEs: Euler Methods and Beyond

Discretization of SDEs: Euler Methods and Beyond Discretization of SDEs: Euler Methods and Beyond 09-26-2006 / PRisMa 2006 Workshop Outline Introduction 1 Introduction Motivation Stochastic Differential Equations 2 The Time Discretization of SDEs Monte-Carlo

More information

Approximate Bayesian Computation

Approximate Bayesian Computation Approximate Bayesian Computation Michael Gutmann https://sites.google.com/site/michaelgutmann University of Helsinki and Aalto University 1st December 2015 Content Two parts: 1. The basics of approximate

More information

Lecture : Probabilistic Machine Learning

Lecture : Probabilistic Machine Learning Lecture : Probabilistic Machine Learning Riashat Islam Reasoning and Learning Lab McGill University September 11, 2018 ML : Many Methods with Many Links Modelling Views of Machine Learning Machine Learning

More information

Part IV: Monte Carlo and nonparametric Bayes

Part IV: Monte Carlo and nonparametric Bayes Part IV: Monte Carlo and nonparametric Bayes Outline Monte Carlo methods Nonparametric Bayesian models Outline Monte Carlo methods Nonparametric Bayesian models The Monte Carlo principle The expectation

More information

Bayesian Inference for DSGE Models. Lawrence J. Christiano

Bayesian Inference for DSGE Models. Lawrence J. Christiano Bayesian Inference for DSGE Models Lawrence J. Christiano Outline State space-observer form. convenient for model estimation and many other things. Bayesian inference Bayes rule. Monte Carlo integation.

More information

Lecture 6: Markov Chain Monte Carlo

Lecture 6: Markov Chain Monte Carlo Lecture 6: Markov Chain Monte Carlo D. Jason Koskinen koskinen@nbi.ku.dk Photo by Howard Jackman University of Copenhagen Advanced Methods in Applied Statistics Feb - Apr 2016 Niels Bohr Institute 2 Outline

More information